open All Channels
seplocked Test Server Feedback
blankseplocked Results of extensive damage/hit testing on Entropy (1618)
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 2 [3]

Author Topic

Sewell
The Happy Spacemen
Posted - 2004.08.02 22:40:00 - [61]
 

Quote:
Estarriol

P. S.: Please call off your lapdogs. That is, those who have nothing to add except 'I have no understanding of the details of this discussion, but i think TornSoul is a swell guy who does great work so he must be right.'


I may not be qualifed to judge which one of you are correct with regards to the facts (no math degree here Rolling Eyes).

But I do know one thing: you, Estarriol, is extraordinary rude, and is acting like a total idiot.

I really don't wan't this to turn into a flame-fest, because I enjoy reading about stuff like this, game mechanichs, nubers etc. But the way you make your case and the way you have replyed to TornSould is totally uncalled for. If you have a diverging opinion, please state it in a clear and concise manner, instead of insulting someone else, it is just lame, and makes you look like a fool.

DeFood
Gallente
UMEC
Posted - 2004.08.02 23:48:00 - [62]
 

Originally by: TornSoul

This is ofc my subjective view, but to me its *very perceptible*, and it's always bugged me....


Ah, Damage is what happens when you *hit* the target.

The randomness of it really doesnt bother me. What bothers me are more the incongruities of the "to hit" model.

Which is, I think where you started when you posted the... uh, exponential dropoff curve - your damage degrade curve.

That curve, really defines, in the ideal case, how both of us would like the to-hit chance to be calculated. But the roll for damage is really a seperate issue. There is no reason to expect that striking targets dead center does more damage than striking them near the edge. Its possibly just as likley that critical components are near the edge of a ship as to the center. So its best to do two rolls. One roll determines if the target is hit or not. Then, if the target IS hit, THEN you roll for damage, using a different distribution to model the chances of the shell hitting a sharply angled surface and glancing off, or hitting some face surface dead on and penetrating for maximum damage.

As it goes, I dont like the fact that:
1. the it becomes 4x harder to hit at double the distance isnt implemted.
2. The guns dont have an accuracy in radians quoted.
3. The speed of the projectile isn't a factor. Shooting at a fast frigate 80km away - scre the "optimal distance" calculation, at great ranges, moving (well, we have predictive computers, so dynamically accellerating targets) should be hard to hit!
4. It bothers me that a lasers damage is not proportional to the power grid or capacitor draw of the weapon. It is energy that does damage, and lasers are simply way of taking an output from a power supply - in megawatts, storing it in a capacitor, and then delivering the accumulated number of joules of energy into a small area on the targets hull.

*thats* what bugs me. The flatness vs the gausianness of the damage distribution? small fries compared to the big glaring lack of rigorous physics issues :P

Selim
Akh'Vehlr Industries
Posted - 2004.08.03 03:29:00 - [63]
 

Edited by: Selim on 03/08/2004 03:31:26
Originally by: dalman
Originally by: Selim
Edited by: Selim on 02/08/2004 02:09:57
The best thing for CCP to do right now would to be give TomB a 'vacation' and hire someone who tries to add content instead of taking away options. Every patch for the past few months has been for taking away options instead of adding content. And even though he is trying to balance ships, he doesnt know a damn thing about what he's doing. Guns are screwed up, cruisers are useless, cruiser guns are useless, most battleship guns are, too, and projectiles were hurt the most. Only missiles are good now. People use frigate neutron blasters on cruisers now, because they are better damage over time than even the best cruiser guns.

I dont think they are even listening to this thread...

Sigh, Selim, stop it.

Plz tell me ONE battleship gun which is useless. Because there is none. Unlike prior to the patches, where there were some useless guns.
And no, small guns does NOT perform better on cruisers than medium guns, except against frigates if you don't have a webbifier fitted (but without one you'll not hit a frigate anyway).
And CCP are working on cruisers. (like the 2:nd bonus for example).



Nevermind. Projectiles are just pathetic nowdays. If they werent, why do you see so many more Ravens and Apocs around now?

I'm just a bit ****ed off at all the nerfs, thats all. The game needs content and actual fixing, not nerfs...

Soren
PAK
Posted - 2004.08.03 05:14:00 - [64]
 

Got your mail to be a BH yet torn?

Excellent work.

GardenerOfEden
Deep Space Fishing
Posted - 2004.08.04 00:59:00 - [65]
 

Fascinating Shocked Let me see if I would expect the same statistical data working with the concepts involved as follows …


DAMAGE:

The damage potential of a projectile is independent of range – explosive energy of warhead remains constant throughout trajectory and while the shell is imbued with additional kinetic energy when fired, that energy is not reduced over distance travelled by atmospheric friction or significant gravitational forces in space.

Before continuing I need to make two points at this juncture. Firstly, an energy weapon would do less damage at range – on the basis that it consumes its own payload over distance. This can also be contrasted with missiles that like projectiles also have a static payload but which, unlike projectiles, can alter their kinetic energy as a consequence using propulsion to manoeuvre onto the target (one way to conceptualise cruise missiles) and by using propulsion to rapidly accelerate towards target (one way to conceptualise torpedoes).

Secondly, effective damage can be scaled up by hitting a vulnerable part of the target in a David vs Goliath “magic bullet” sense. However, this is really a matter that concerns the target itself and where it is hit. While range influences accuracy, which in turn influences where the target is hit, it does not change the damage of the shot as such; accuracy merely changes the consequences of that inherent damage.


GAME MECHANICS: The Mechanism

What I would have done in terms of combat mechanics at this point is to process each shot by taking a snap shot of the target’s range and transversal velocity at the time the weapon is fired and calculate what amount of tracking would be necessary to hit the target. Note target size is irrelevant in this context as that merely influences the ease with which those parameters can be obtained with scanning sensors albeit that one would expect smaller frigate like targets to be faster moving i.e. higher transversal velocity potential.

If the tracking that was required is within the parameters of the weapon adjusted by the effect of tracking modules and the Motion Prediction skill then treat the shot as a hit and process damage, which may at that point involve considering the randomised chance of inflicting special/critical damage (note this is the only time a random number would be utilised) and would be influenced by the Surgical Strike skill.

In order to add a bit more realism, and lessen the harsh extremes of an all or nothing hit or miss, what I would do is check in the case of miss if the miss was within a slightly enhanced range of tracking (e.g. the weapon’s adjusted parameters plus a percentage increase influenced by the target’s actual physical size) and if so award a “consolation prize” of the damage reduced in proportion to the extent of the miss to simulate a partial hit / glancing blow / proximity detonation (e.g. the weapon’s parameters + 5% = ˝ damage though this does not have to be linear and actually should not be linear from a realism perspective). There would be no chance of doing inflicting special/critical damage with this consolation damage.

Note when calculating damage it would also be possible to adjust it for weapon / ammunition type (e.g. an explosive shell might still do a lot of consolation damage in contrast to an energy weapon that really needs to actually hit in order to damage) and range (e.g. if it is accepted that an energy weapon consumes its own payload then damage should be reduced in proportion to distance of target from weapon).


GAME MECHANICS: The Problem

A problem exists at this point with determining whether a shot hits or misses at the time it is fired as the shell still needs to travel along its trajectory and impact the target. Being able to align the weapon at the point fired at is only part of the equation. The other part is the target behaving as predicted so as to actually be at the point aimed for at the exact time the shell arrives such that the shell impacts the target. This is harder to do with a more agile target, which gives it more potential to act unexpectedly, and harder to do if the target is at a greater distance, which gives it more time to act unexpectedly. Hence some form of adjustment is required for the sake of realism and because otherwise a ship with a weapon that does a lot of damage that can positioned at a distance so that it is easier to track targets but which hits the target instanteously might be too powerful – oops anybody who isn’t a Tempest captain with 1400mm howitzers would have used the word “would” instead of “might” sorry Embarassed


GardenerOfEden
Deep Space Fishing
Posted - 2004.08.04 01:00:00 - [66]
 

Edited by: GardenerOfEden on 04/08/2004 01:02:48


CCP used to make this adjustment through the use of optimum range and falloff range – once the target got passed optimum then damage was scaled by the proportion of falloff that remained e.g. go 20% into falloff and do 80% damage, go to the end of range at optimum + falloff and do 0 damage. What this means is that you could position your sniper ship at long distance where it could hit but the reduced damage compromises the tactic. The “sniper” therefore has to come closer where it is harder to track the target and where the likelihood of the target getting the better of the sniper is increased i.e. game balance is restored.

If the damage from your test scenario was plotted on a graph at this point then all that that statistical data should do is confirm constant damage to optimal range and then the linear reduction in damage as range moves outwards to optimal + falloff if special/critical damage of wrecking shots is disregarded as a statistical outlier or the 1/100 chance of such damage occurring if it is included. Note the damage is constant because there are no misses as no tracking is involved with a stationery weapon and stationery target.

There would be a lot of players who would be very happy these mechanics; however, I would not be one of them. This is because of the amount of damage dealt reduces over distance, which is physically incorrect as I outlined initially and totally contrary to the conceptualisation of a howitzer, which pulverises its targets at long range.


GAME MECHANICS: The Solutions

There are four options that use the above mechanics but make an adjustment for the above problem and they all involve resolving the problem by reducing tracking accuracy over distance in contrast to reducing damage over distance:

1. Model Reality – in theory the projectile shell could be placed in the 3D game environment and moved along its trajectory to determine if a hit actually occurs just like a missile, which would also allow contact with other intervening ships and obstacles like asteroids, to be addressed. However, CCP has obviously decided that it is not practical to calculate a trajectory for the shell and process its straight line flight path in a MMORPG so this option must be rejected and there is no need to consider how it would look on a graph;

2. Reduce Tracking Linearly – this keeps things simple by still using the adjusted tracking parameter of the weapon but reducing it as the distance to target extends beyond optimal into falloff such that at maximum range of optimal plus falloff you have zero tracking and you can only hit stationery targets and targets that come straight at you i.e. n00bs and NPCs, which have already been flagged for AI enhancement. This is the solution I personally favour for what it is worth. If this mechanic was used and the damage from your test scenario was plotted on a graph then the damage would remain constant over all ranges simply because damage would be independent to range and all shots would hit with stationery targets – the only deviation would be special/critical damage occurring 1/100 shots but again independently to range;

3. Apply a Formula – this is the same as the second option only instead of a linear reduction in accuracy a formula using optimal and falloff ranges and some form of constant value is applied to it to determine effective accuracy. I do not like this for three reasons:

3.1 the best tracking would be at the optimal range, which means that tracking would be penalised as the target moves closer – this is unrealistic as the weapon still has the same basic ability to move and the target has less time to react unpredictably (i.e. the reverse of what we are trying to adjust for from a realism perspective) and is effectively a double penalty because the closer target is likely to have a higher transversal velocity, which means it is harder to track/hit/damage anyway;

3.2 it destroys game play because it means that there will be sweet spots for weapons that would be impossible to achieve in a dogfight and because the change in accuracy would not be linear (i.e. plotted graph of tracking accuracy over distance changes gradient), it would not be intuitive how adjusting range by the same increment of, say, 500m would affect accuracy/hits/damage as it all depends upon the range you are at to begin with; and

3.3 it destroys game balance because given the first two points why would a player choose to (use a ship optimised for) projectile weapons over (a ship optimised for) missiles which are not subject to such vagaries?


GardenerOfEden
Deep Space Fishing
Posted - 2004.08.04 01:00:00 - [67]
 


If this mechanic was used and the damage from your test scenario was plotted on a graph then the damage would also remain constant for the same reasons – the graph would only look different to the linear graph for a non-stationery target when the differences in tracking/hits/damage became apparent;

4. Apply a Probability – tracking could be adjusted by applying probability distribution i.e. a variable factor whereby you have close to 100% of the weapons adjusted tracking at optimal range. This is equivalent to rolling a d20 hit die and needing to roll 19 or less to get good tracking at optimal range but needing to roll 2 or less to get good tracking at max range. The three problems with this type of approach are:

4.1 nonsensical results are achieved such missing a stationery target with a stationery weapon (assuming no outside interference which is the only real way to rationalise such a miss);

4.2 outcomes to do not conform with probabilities e.g. in a short fight using a weapon with a low rate of fire it is unlikely that the actual damage would equate to the probable damage so a statistically improbable outcome can eventuate and the existence of such anomalies is not conducive to game balance and therefore is harmful to game play; and

4.3 randomising hits is usually done to approximate the reality of not always hitting and not always missing, however, given the game mechanics I have outlined above there is no need to do this because you can determine the accuracy of the hits directly –simulating accuracy with the equivalent of a dice throw is redundant and when using those mechanics would act as a double penalty.

If this mechanic was used and the damage from your test scenario was plotted on a graph then the damage would not be constant at any range simply because the accuracy of the weapon is not constant to begin with. However, the amount of damage would correlate to the probability distribution applied to the accuracy of the weapon, which is what I understand your statistical data to not only prove but to also identify. Furthermore, what your statistical analysis also illustrates is how woeful damage becomes at distance i.e. the end result of the mechanics that CCP is actually using is that poor accuracy tracking over longer distances is severely penalised beyond what is realistic.


In conclusion, I consider your statistical analysis to support my conceptual analysis that CCP is using flawed game mechanics to determine accuracy/hits/damage of turrets. I consider my preferred game mechanic outlined above to be superior however I retain an open mind to anyone convincing me otherwise.


Selim
Akh'Vehlr Industries
Posted - 2004.08.04 04:28:00 - [68]
 

Edited by: Selim on 04/08/2004 04:32:49
Wow Gardener. Shocked I hope the devs read that...

As for being harder to hit up close, I completely agree this is flawed, at least against ships of the same size or one down. A gun would still be able to anticipate where a ship is going. In fact, much, much more so at close range. The ship itself is far larger a target up close so its alot easier to it. Against a ship of the same size its literally like taking a shot at the broad side of a barn..

As for projectiles having unlimited range - that would be pretty unbalanced... unless of course their accuracy at longer ranges was very poor, while medium and short range it was good.

DeFood
Gallente
UMEC
Posted - 2004.08.04 07:47:00 - [69]
 

Selim, Infinite range projectiles would not be a problem as long as two things were modelled:

1. Projectile velocity meaning that, one fired there is significant latency before the shot lands up 100km away. Well, lasers dont have this problem I suppose, with a projectile velocity of 300,000 km/s they wouldnt have a problem getting to great range particularly fast.

2. A 4X decreased chance to hit each time the distance to the target doubles.

Selim
Akh'Vehlr Industries
Posted - 2004.08.04 10:02:00 - [70]
 

Edited by: Selim on 04/08/2004 10:05:28
But it would make all ammos besides EMP, phased plasma, and fusion obsolete. Although they could just have the ammos do variations with the different damage types...

They'd have to make the tracking alot better on the projectile guns, though, so they could hit really good at medium ranges.

But what about autocannons? They'd have to have a bigger penalty for range.

But I do think the idea sounds good... but it would be pretty hard to implement. Regardless, projjies need some love(tm)

It would be a great idea also, because anything that diversifies the different weapons is good in my book. Projectiles having your proposed new system, being good at short and medium but can hit at extreme range, but poor accuracy. hybrids staying the way they are, lasers staying the way they are. And missiles of course are different enough. Wink

Tar om
Minmatar
Octavian Vanguard
Posted - 2004.08.04 11:31:00 - [71]
 

Tracking and RoF

While we are (loosely) on the subject of improving realism in EVE gunnery, can we consider the way that turrets currently fire at their maximum RoF in all situations? It seems that the guns are on full auto, and are forced to fire as soon as they are loaded (or charged, or cooled, or whatever). This strikes me as overly simplistic and the source of several inconsistencies in the game.

To put it simply, max cyclic rate of fire does not equal the effective rate of fire. The reason? Aiming. It seems that the turrets are on commission for every round fired, which means they fire as soon as the turret is able nomatter what the chance of hitting the target is. Surely it would be more sensible for turrets to wait until they have a decent firing solution before letting rip with their next round? As EVE works now this does not happen so your chances of hitting a target are vastly reduced if the target is attempting to evade simply because you only get the chance to hit the target ever x seconds according to your RoF _and are forced to take that chance _ instead of waiting for your best shot then reloading and trying again.*

Directed gunfire is nothing new in a naval context (it was present at Jutland) if we could have this modelled in EVE then it would add massively to the tactics available in the game. How could this be done?

I would like to see Firing Rules for turrets. If you could (optionally) set the firing parameters that would have to be met before your guns fire, we could then manouvre our ships to get into an advantageous firing position and have the guns fire when they bear instead of manouvring the ship and holding position until a gun happens to reload. My initial guess on parameters would be Range (max and min), Transverse Velocity and Target Signature radius. There could be a lot of fun to be had with the various targeting computers (most of which are only fit for scrap at the moment) if some of them allowed extra parameters, more or less solutions or the very top computers might allow CCIP style tracking.

OK, its complex. But I don't think that the current gunnery system does the rest of EVE justice.




*I get the feeling that this is the root cause for 1400mm users' unhappiness with the current situation, since testing for DoT all seems to be done while stationary.


Ray McCormack
Nordar Innovations.
Posted - 2004.08.04 12:50:00 - [72]
 

Originally by: Estarriol
P. S.: Please call off your lapdogs. That is, those who have nothing to add except 'I have no understanding of the details of this discussion, but i think TornSoul is a swell guy who does great work so he must be right.'


You have unfortunately lost all credibility with me here. I was at one stage interested to see where your dicussion with my CEO would go, as I'm sure he was. But your petty labelling of others in this thread dumbfounds me, and I can only assume that what is left of your bruised ego is indeed 'looking to pick a fight'. While I may be in no position to address you with regards to the discussion on this thread, your casual reference to the average poster as a miniature beast of pleasure has entitled me to engage your obviously disdainful scorn.

Your argument, while bordering on almost juvenile in method now, is constructive. Your out-of-hand mockery is however not. Please refrain from it. To someone such as yourself, it should indeed be beneath you.

TornSoul
BIG
Gentlemen's Agreement
Posted - 2004.08.04 17:47:00 - [73]
 

Originally by: DeFood

Ah, Damage is what happens when you *hit* the target.

The randomness of it really doesnt bother me. What bothers me are more the incongruities of the "to hit" model.



It's a matter of how you look at it really.
Ie. what 'model' you choose to use.

I prefer to think of a miss, as just another hit, but with zero damage. If you see what I mean.

Thinking like this simplifies a few things (the way I think about it anyhow).

To ilustrate this - Let me use an analogy *that has nothing to do with ingame stuff*, but merely servers to illustrate this point and nothing else (gotta love disclamers..)

At a certain range you simply roll a dice - and get a value from 1 to 6. Which is then the damage you inflict.
At a further range, where its harder to hit (and you thus get a number of misses) you do the same roll, but subtract 2 from the result.
You then have a range of -1 to 4.
Anything below 1 is simply threated as a miss.
(despite my disclamer - this is actually close to how it works ingame atm)

No need for two calculations (hit+dam)- it's all done in 'one go'.

One problem is ofc that you no longer can get damage above 4.

Introducing Normal (gaussian) Distribution:
The range of possible values from *any* Gaussian distributon is all the real numbers - from minus infinity to infinity.
As before we simply agree that any value below (in this case) zero, is simply a miss.
Other values - are the damage inflicted.

By simply adjusting the Mean value, say reduce it, we in effect do the 'subtract 2 bit' (from the dice example) at greater ranges.
But we *still* have the possibility of doing damage of '6' (the max with the dice from before) - so we havent 'cut of' any values.
What we have done instead, is (greatly if we wish) reduce the chance of high damage happening (but not eliminate it)

So : By simply reducing the Mean, we increase the number of misses (which is what we want at higher ranges) (assuming we use a 'cut of point' at zero)

Using Gaussian distribution, we have one further option we can 'play with' - Standard Deviation(SD).
Basically : The higher SD, the more 'random' will the shots be. (SD = inifinity = uniform distribution, where each value(damage) is equally likely to happen)

A higher SD at longer ranges is (in additon to a reduced Mean) also what we want, as the shots will be more 'spread out' (closer to random) at long ranges.

Now here usually comes the argument about the ship not beeing most vulnerable at its center.
Please - I'm saying no such thing (as Ive said a few times already).
For 'RP reasons': Assume our targettting computer knows where all the vulnerable 'spots' are (the ship designs are afterall not really any secret) - and thus alwasy tries to aim for one of them.
As range increases, it becomes increasingly harder to hit one of those - ie. a larger SD.
If there is one or many vulnerable spots really doesnt matter a bit.
It's the *ratio* (area if you will...) between the vulnerable spots and the more armored spots that matters.
Not their actual position or anything else.


@GardenerOfEden
Wow - Thats one mouthfull.

One thing I disagree with - purely from a gameplay perspective - is the notion of shells always doing the same damage regardless of range *if* they hit.

While physically this is correct - I think (prefer) that in a gaming environment this should not be the case.
As gamers we are 'accustomed' to that 'long range = less damage' (wheter it beeing individual hits or damage over time).
Again - This has nothing to do with RL physics, but is strictly a (my) gaming preference.
Each to his ofc Wink

FileCop AI
Amarr
Finite Horizon
Posted - 2004.08.04 18:18:00 - [74]
 

Just to pop in some negativity:

Why change what isn't broken?

Aren't we all shooting away and having fun? Yes..

Why mess with **** then? And I like how you say it "should" be like this and that.. Did you program the game? No. Then you don't know how it "should" be.


TornSoul
BIG
Gentlemen's Agreement
Posted - 2004.08.04 18:34:00 - [75]
 

Fair enough - But at least read the thread before commenting.


I already said that
Quote:

I'm not suggesting CCP immediatly drops everything and fixes this (theres far more important matters to attend to atm) - The current system *does* work afterall
I'd just like to improve on it is all - and this is my proposal to do just that.


Valorian
Val KorP
Posted - 2004.08.04 18:47:00 - [76]
 

Edited by: Valorian on 04/08/2004 18:49:58
Originally by: dalman
Originally by: Grim Vandal
could you do the same for the 1400mm howitzer I? Embarassed

Since I heared that the 1400mm hits way better at optimal + falloff (not sure if its ture since I cant use them myself but I want them to be balanced anyway)

That post is excellent, kudos to you.



A very large test was done some months ago in pretty much exact the same way as this one (but with even more data and with different turrets). And it showed what everyone would think: ALL GUNS USE THE SAME FORMULA.

So the graphs of a 1400mm is exactly the same, you just need to stretch it out to the optimal and falloff of a 1400.

What people are complaining about on 1400 is tracking...
Hits depends on 2 variables, range and tracking, and this test, as most others, are made with tracking as a fixed value (perfect hits).
A test with tracking can easily be done by orbiting the secure can, but would require much more data and work (due to different orbit speeds at all the different distances).

And btw, most people whining about 1400mm are just, errr, whining ;)
Tracking on them is nowhere near as "bad" as they make it look.



Actually with 1400mm & 1200mm & 800mm, within (about) half the optimal range and irrespective of tracking (ie the target is flying straight at you), misses are what occur every shot. This is an observation after using several thousand rounds. The same appears to happen with Medium and Small size projectile guns. (On Tranq)

Celt Eireson
Minmatar
Posted - 2004.08.04 23:04:00 - [77]
 

Hmm may have been mentioned earlier in the thread and I missed it but just in case :-)

Whilst not understanding all the statistical discussion, I would assume that to make the weapon calculations more realistic would require somewhat more complicated calculations? I'd also assume that the more realistic and hence more complicated the calculations the bigger hit this would have on the servers? I know the calculation for one gun takes only a very minute amount of processor time, but that's for every gun firing (both PC and NPC?) every time it fires for every ship in combat at any one time, it does all add up.

GardenerOfEden
Deep Space Fishing
Posted - 2004.08.05 12:02:00 - [78]
 

Originally by: Selim
But it would make all ammos besides EMP, phased plasma, and fusion obsolete. Although they could just have the ammos do variations with the different damage types...
Yes the different damage types plus damage radius - in a longer range scenario you might opt to use more of a proxmity explosion type of shell (i.e. less proper damage but more consolation damage under my mechanics) given that the target would be harder to hit if it was taking evasive action.


GardenerOfEden
Deep Space Fishing
Posted - 2004.08.05 12:22:00 - [79]
 

Originally by: TornSoul
@GardenerOfEden ... One thing I disagree with - purely from a gameplay perspective - is the notion of shells always doing the same damage regardless of range *if* they hit.

While physically this is correct - I think (prefer) that in a gaming environment this should not be the case.
As gamers we are 'accustomed' to that 'long range = less damage' (wheter it beeing individual hits or damage over time).
Again - This has nothing to do with RL physics, but is strictly a (my) gaming preference.
Each to his ofc Wink
In game it wouldn't seem as if the shells were hitting for the same amount of damage because of the randomised effect of the damage on the target and because movement would mean that a lot of damage would be what I have termed "consolation" which would vary given it would be proportional to the extent of the miss.

Movement would also reduce effective accuracy at long range so you would get the drop off in DoT that you want, however, if the target doesn't move very well then it would get damaged instead of being able to ignore a long range threat (which is what I want to see happen). Don't forget that a missile suffers no diminution in damage with range - sure it takes longer for the inital missiles fired to impact the target but once the missile spam starts hitting the rate of impact approximately equals rate of fire of the launcher.


DeFood
Gallente
UMEC
Posted - 2004.08.05 16:52:00 - [80]
 

Originally by: TornSoul
It's a matter of how you look at it really.
Ie. what 'model' you choose to use.

I prefer to think of a miss, as just another hit, but with zero damage. If you see what I mean.

Thinking like this simplifies a few things (the way I think about it anyhow).


I dont think theres much merit in swapping one simplified system,for another simplified system.

I like to think of misses as, well, outright misses. Then its easy to see why the target wasnt damaged. Because the projectile actually didn't hit it at all.

Seperating the hit and damage rolls allows one to use two different curves... Which makes a lot of sense because, frequently they *need* to be different.

Lasers, projectiles and hybrid weapons could/would then have (potentially) very different tracking and damage characteristics.

Its not clear at all why - for instance - we would want to reduce the damage for projectile or hybrid weapons at range. It is unlikley that their velocity is reduced - this is space afterall with 0 friction - additionally the projectiles at least contain armor piercing or are explosive - theres no reason at all to reduce explosive damage at range. As for the hybrids, perhaps the plasma cools?

Lasers? depending on their focusing the beam will "spread", decreasing the energy density of the beam. The same amount of energy spread over a larger area will simply cause heating of the ships armor rather than damage.

Additionally, the same drag force that limits ship speeds could be modeleld for projectiles and hybrid weapons :- that though would give a totally different curve than we currently have - and would influence the kinetic component of the damage produced. Explosive would remain constant, thermal damage on projectiles should remain constant, on hybrid I expect plasma to cool fast.

Uh, where was I going with this. Ah yes. Your system, as it stands now, is just as much a simplifcation as CCPs. Neither system maps to "reality" any better than the other. As Such I dont see the value in merging the hit and damage calculations.

Keeping them seperate gives the developers more opportunities to make weapons balanced but different, which I find interesting. And if there is one thing that computers are good at its the sorts of calculations involved in doing damage and hit rolls :P So theres nothing to be gained - performance wise, by simplyfying the rolls.


TornSoul
BIG
Gentlemen's Agreement
Posted - 2004.08.05 17:13:00 - [81]
 

Originally by: DeFood
Keeping them seperate gives the developers more opportunities to make weapons balanced but different, which I find interesting.


Fair enough point - and one I'm not adverse to either.
I simply prefer the 'mathematical beauty' of having it all in one 'formula' - Which admittedly is not a very strong argument Wink

Grim Vandal
Burn Proof
Posted - 2004.08.12 14:26:00 - [82]
 

EVE-I has a great tool now for analyzing combat data:

Link

If someone wants to compare 1400mm with tachs and 425mm rails Very Happy.

would really like to see the difference IN PURE DOT!!!

Looking forward to your testing results Razz

ugh

Skyt Inteky
Posted - 2004.09.10 19:32:00 - [83]
 

But what about the log ??

It doestnt work well, many shots are not logged...

submit petitions about that

Imhotep Khem
Minmatar
Doom Guard
Wildly Inappropriate.
Posted - 2004.09.20 02:43:00 - [84]
 

I think all the statistics cause people to loose sight of TornSoul's point #1.

By classifying, and naming the hits, CCP implied bell curve distribution. But, the classification is false because each 'class' of hit is the same.


I think all the harsh comments cause people to loose sight of Estarriol's main point. If you ignore the classifications, it all makes sense.

Conclusion: CCP, either give bell curve distribution for the hit classes, or remove them.Razz

Loka
Gallente
adeptus gattacus
Lotka Volterra
Posted - 2004.09.20 09:27:00 - [85]
 

Originally by: Skyt Inteky
But what about the log ??

It doestnt work well, many shots are not logged...

submit petitions about that


If you fire only one weapon, all shots will be logged. The problem of not logging hits/misses are coming with lag and weapons hitting the same time.

FireFoxx80
Caldari
Caldari Provisions
Posted - 2006.05.15 16:04:00 - [86]
 

I hate to necro. But is the maths here still valid? Has anyonne bothered testing?

Jorauk


The Scope
Posted - 2006.05.15 16:55:00 - [87]
 

Originally by: FireFoxx80
I hate to necro. But is the maths here still valid? Has anyonne bothered testing?


Then please don't. In the future, make a new post regarding your question, perhaps linking to this one for further discussion.

Locked for necro.

*Click*


Pages: 1 2 [3]

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only